Background: Family and twin studies indicate that panic disorder is a genetic disease. These data indicate that the genes could act either independently, by either a dominant or a recessive model, or additively. We will complete a 10 cM exclusion map of panic disorder by the end of the current project period. This will test the hypothesis that a major locus accounts for panic disorder in as many as half of the pedigrees. If a major locus is not identified by us or other groups working on panic disorder, it is unlikely that one locus win account for panic disorder, and greater statistical power will be needed to detect disease genes. Power can be increased by expanding the phenotype, increasing the number of pedigrees, and in making the marker map denser. The specific aims for the next project period address these needs. Progress: We have collected 23 panic disorder pedigrees with 90 cases of DSM III- R panic disorder. We have typed over 400 DNA markers in these families and obtained lod scores less than -2.00 over 85% of the genome under both dominant and recessive models of inheritance. This work has identified a region on chromosome 7p12 where the maximum lod score reaches 2.76. Studies of candidate genes have found a mutation in a transcription factor 5p 1 site of the CCK gene that is present in 34% of panic disorder patients compared with 16% of population controls. These findings need to be followed up. Specific Aims: In response to the last review, the principal aim of the next grant period is to create a 10 cM joint exclusion map of panic disorder by combining our pedigrees with those of our collaborators at Columbia University. Other aims are to create 1 cM maps of all regions of interest in the Iowa pedigrees, add three linkage pedigrees per year, perform CO2 inhalation tests for panic attacks on existing and new pedigrees, and collect sample of subjects with panic disorder for an association study.